GB2101346A - Testing film transport mechanism - Google Patents

Abstract

Cine or video film transport mechanism is tested for picture steadiness by producing a first colour component of an output image and comparing with a second contrasting colour component of the output image as reference. The colours may be green and magenta. In a printer, an unexposed positive film may be exposed through a negative carrying a two colour test image, in a first run to one colour and in a second run to the other. In a camera, film may be exposed to a test image of two colours in a first run using a filter of one colour and in a second run using a filter of the other colour. A projector may be operated (a) as a camera (without the projector light) or (b) to project a one colour component of a test image and compare with the other colour component of the test image as a fixed image. The test image may be a Vernier scale or a "flat stair-case" (as shown). Testing may be effected automatically by comparing the white and black regions of the output image. <IMAGE>

Description

SPECIFICATION Picture unsteadiness in cinematographic film transport mechanisms This invention relates to the measurement of picture unsteadiness in cinematographic film transport mechanisms. These may include cine cameras, printers and projectors and also telecine machines.

A major problem in measuring picture unsteadiness associated with cine film residues in the difficulty of isolating the various different causes of unsteadiness. For example, it is difficult to isolate variations caused by variations in the physical dimensions of the film stock within its manufacturing tolerances, from variations caused by the film transport mechanism under consideration.

Many methods have been proposed for identifying and measuring the contributions to total picture unsteadiness which are made by the various individual mechanisms through which film passes in the process of transferring the image from the original scene to the screen, but none satisfy all the requirements.

In general it can be said that satisfactory methods exist for testing the accuracy of intermittent film transport mechanisms. This is normally achieved by exposing a piece of film to a suitable test object during each of two or more passes, such that any interaction between the parts of the resultant composite image is a direct indication of the inability of the mechanism to position the film precisely. Within reasonable limits such indication is independent of film tolerances. Clearly the same principle can be applied to any intermittent mechanism, whether it be part of a camera, printer or projector.

Our British Patent GB-PS 90071 8 (Inventor: A. Branson) describes a method using a vernier scale on a static test chart. To detect unsteadiness in a camera the unexposed film stock is passed twice through the camera. On each pass the film is exposed to one of the two respective halves of the vernier scale. The other half is mechanically screened. Movement of the film between the two passes will then be detectable by inspecting the positions of the images of the two halves of the vernier scale on the processed film. A problem with this is that moving the mechanical screen can cause movement of the whole vernier scale.

No direct method then exists to measure printer unsteadiness. The problem with a continuous printer is different, for the reason that such a printer is designed to transfer information usually by contact from one strip of film to another continuously, at high speed. A continuous printer is described in British Patent 1,200,612.

The test information therefore cannot be obtained direct from a static test chart, it must be recorded on a representative strip of film which is used like a normal negative. This inevitably creates the problem of isolating the printer's contribution to unsteadiness, from the inherent and inevitable variations which are associated with the negative and positive films, as a result of manufacturing tolerances.

Printer unsteadiness is a particularly important factor, as has been noted for example in a paper by J.R. Sanders in British Kinematography Sound and Television November 1971, at Sections 5.6 and 6 on pages 399 400. The method of determining printer unsteadiness described in our Patent 900718 involves estimating the combined effect of camera and printer unsteadiness and then making an allowance for the camera unsteadiness.

The present invention in its various aspects is described in the appended claims to which reference should now be made.

In a preferred embodiment of the invention a strip of colour film (which for the purpose of this description will be called "the negative") is exposed in such a way as to produce an image, one part or component of which can be illuminated in one pass through the printer to expose the corresponding part of the positive film and another part or component of which can be illuminated during a second pass through the printer to expose another part of the film. When the print is viewed by projection, the interaction between the two elements of the composite image provides an indication, independent of the slitting and perforating tolerances of the film, of the inability of the continuous printer to position the negative with respect to the positive in the same position on the two successive passes through the printer. An analogous system can be applied to a camera.

The invention will be further described with reference to exemplary embodiments thereof.

Reference will be made to the drawings, in which: Figure 1 shows a preferred new test pattern of "flat staircase" form; Figure 2 shows a negative form of test pattern; Figure 3 illustrates the incorporation of the test pattern into an existing pattern; and Figure 4 is a block diagram of circuitry for automatically monitoring the film to provide an indication of unsteadiness.

In order to test a printer, a negative is first made from colour film stock. The test object from which the negative is made could take many forms. One example uses the Vernier scale test object described in British Patent Specification 90071 8. For this particular purpose however one half of each pair of Vernier scales is exposed to magenta light and the other to green light in one pass through the camera, thereby providing a composite image on the negative which is the complimentary of those two colours.

The method used to test the printer is as follows: (a) Identify a specific frame both on the negative obtained as above and on the unexposed colour positive (any frame will serve and a good method is to punch a hole), to ensure that they can be placed in coincidence at the beginning of each of two runs through the printer.

(b) Locate the two identified frames on a given tooth on the master printer sprocket.

(c) Expose the film during the first run to magenta light.

(d) Rewind the two films to the start mark (without exposure).

(e) Replace both films as described above and expose the second run to green light.

Normal processing will produce a composite image of the Vernier scales with one half of each pair of scales being green and the other half magenta. Since the same two frames were coincident and located by the same tooth for both passes through the printer, the resultant interaction between the two images is an accurate indication of the inability of the printer to position the two strips of film in the same place twice during exposure.

Having established the performance of the printer, the same negative can be used to obtain further information. A further length of colour positive is exposed to the first run as described above, but for the second run the negative is displaced with respect to both the positive and the chosen sprocket tooth. The resultant interaction between the two images will indicate the magnitude of the combined tolerances in the pitch and position of the perforation holes. This gives information about the quality and stability of the film stock. Any colour capable of sufficient separation from another could be used if such a combination proved more suitable for a particular application.

Using the Vernier scales it is possible to make an assessment, independent of the variations in the pitch and the position of perforations on the film, of the performance of continuous printers.

This is realised by viewing the interaction of the two images whilst the film is being projected normally.

The Vernier scales are convenient for continuous subjective assessment by projection.

They are not ideal if a large number of frame-byframe measurements are needed. For this purpose we propose a new test pattern which is more suited for the continuous measurement and recording of frame-by-frame variations which occur in double exposure tests both from the film camera and the printer.

The new test object is a transparency and avoids mechanical shutters or flaps by using the two colour method previously described for producing double exposure camera and printer tests. Figure 1 shows the proposed new test pattern which takes the form of a "flat staircase" with each step increasing in height by say 0.1% of picture height (or width). The number of steps corresponds to the required range, in this example +0.3%. A mirror image of the staircase is inverted beneath it to form a symmetrical stepped core of one colour surrounded by, but separated from, a generally similar stepped pattern of the opposite sense and of contrasting colour, such that chain rectangular spaces are left between the steps. For test purposes magenta and green have been used but other combinations may be more suitable.

If such a "positive" type pattern in the form of a transparency is back lit, part of the film in the camera is exposed to magenta light, part to green light and the clear rectangular spaces is between the steps to white light. These rectangles will reproduce black on the negative and clear on the resultant print. The image pattern on the negative is produced with a single run through the camera, and when printed twice as described above, is suitable for deriving positional information related to the printer unsteadiness and which could be handled by an automated method.

When used as a double exposure test for camera, "negative" type test pattern (Figure 2) is arranged to produce opaque rectangular spaces between the "stepped" patterns which will then be clear on the negative, rendering it suitable for the measurement of camera unsteadiness with the same automated apparatus. In this case, one run through the camera is exposed with a magenta filter over the lens, and a second run with a green filter over the lens, so that each run produces a separate image by transmitting light which is the same colour and "rejecting" light of the complimentary colour. The opaque rectangles positioned between the two stepped patterns will reduce the transmitted light to a level where no effective exposure occurs.In practice, both negative and positive patterns are combined on one transparency and patterns can be arranged to indicate both vertical and horizontal movement.

If no interaction is present between the two coloured images on the film, the clear rectangles will remain unchanged in size and relative position. If a small amount of movement in either direction occurs between the two images, i.e. the core and the surround, the clear rectangles must decrease in "height" on one side of the core and increase by a corresponding amount on the other side. When the excursion is great enough to obscure either of the smallest rectangular spaces, it can be said that a relative movement equal to or greater than 0.1% of picture height occurred on that frame. If the second rectangle is also obscured, the movement must have been 20.2% and so on. Since the position of the core with respect to the surround will vary both in magnitude and direction, the clear rectangle will sometimes be obscured on one side of the core and sometimes on the other. This change in position can be interpreted as "positive and negative going" information. Since each frame contains what amounts to binary information (i.e.

rectangles obscured or not obscured), it becomes possible to design a device to read and record this data.

Figure 4 illustrates apparatus for providing an automatic determination of movement. Ten sensors 10 of linear shape are arranged under the ten green/magenta junctions that occur in the flat staircase test object. It will be appreciated that in this case there is one further step than shown in Figure 1. These sensors ignore the green and magenta components but are able to detect any white or black region of their junction. With an image generally of the form of Figure 1, black regions will be formed if there is displacement of the green and magenta areas such as to cause them to overlap.Each sensor is associated with a respective white level comparator 12 and black level comparator 14, both of which receives a reference input from a reference (black) patch detector 1 6. Error check logic 1 8 checks the outputs of the twenty comparators for selfconsistency, so as to eliminate errors caused by dirt on the film for example.An appropriate output is then provided indicating the magnitude of the error (within ranges defined by +0.4% of the picture height or width) and this can be applied to an analogue or digital output as shown.

As an alternative to having twenty sets of comparators, it would be possible to use a single pair of comparators and apply the outputs of the twenty detectors to them sequentially.

The system can desirably be used with a colour tri-pack film, the individual layers of which are used to record the two elements of the double exposure test.

Long experience indicates that the Vernier scale is instantly understood and totally acceptable for the day-to-day subjective assessment of tests made by technicians who are responsible for the maintenance of cameras and film handling machines. It is equally known from long experience, that an automatic method of measuring and recording frame-by-frame excursions is essential to permit a detailed statistical analysis of the performance of the film system.

It seems unlikely that one test pattern will be optimum for all requirements but using the proposed two-colour double exposure technique, thereby avoiding mechanical devices, it is possible to combine in a single transparency the desirable features of other methods. Figure 3 combines the stepped test pattern with established Vernier scale, so-called "sharks teeth", and stripe pattern designed for measuring total picture unsteadiness using a telecine in conjunction with a special meter (or oscilloscope).

Such a test chart would be comparatively easy to make and subsequently to reproduce.

As described above, therefore, a printer is assessed by printing onto a positive a two-colour measuring scale, one of which may conveniently be regarded as a reference and the other of which will show any relative movement. Likewise a camera is assessed by exposing onto film stock a two-colour image for assessment of unsteadiness. A projector can be assessed in one of two ways. It can be used in reverse as a camera, and loaded with unexposed positive fi-lm.

A static test scale is then imaged into the film using the projector optics in reverse; naturally the projector lamp is removed. This is found in practice to produce acceptable results despite the fact that there is no shutter between the image and the film. Alternatively, a processed film exposed to half of the vernier scale can be projected. The image projected from the film is of one of two contrasting colours, either by use of a filter or because the scale on the film is coloured.

A beam-combining arrangement then combines this with another fixed scale of the contrasting colour. Unsteadiness of the projector film transport mechanism manifests itself as movement between the two scale parts of the displayed image.

Claims (8)

Claims

1. A method of testing the operation of a cinematographic film transport mechanism having an optical system which provides an output image, in which a first colour component of the output image is obtained during a pass of the film through the mechanism and is compared with a second reference component of the image which is of contrasting colour.

2. A method according to claim 1, in which the second component of the image is obtained during a second pass of the film through the mechanism.

3. A method of testing a cinematographic printer, comprising preparing a negative film stock having a measuring scale with two parts of contrasting colours, running the negative and an unexposed positive film through the printer exposed to light of one of the two colours, running the negative and the positive film through the printer exposed to light of the other colour, processing the positive film, and assessing the apparent movement of the two scale parts on the processed positive film.

4. A method of testing the operation of film stock in a film transport mechanism, comprising preparing a negative film stock having a measuring scale with two parts of contrasting colours, running the negative and an unexposed positive film through a cinematographic printer exposed to light of one of the two colours, relatively displacing the negative and positive films, running the negative and the positive film through the printer exposed to light of the other colour, processing the positive film, and assessing the apparent movement of the two scale parts on the processed positive film.

5. A method of testing a cinematographic camera, comprising providing a measuring scale with two parts of contrasting colours, running an unexposed film through the camera with only one colour component reaching the film, running the same film through the camera with the other colour component only reaching the film, processing the film, and assessing the apparent movement of the two scale parts on the processed film.

6. A measuring scale for use in the method of any preceding claim, comprising two regions of contrasting colours, the two regions having opposed portions separated by white or black regions, the distance between the opposed portions varying along the lengths thereof.

7. Apparatus for assessing the unsteadiness of a film transport mechanism used in obtaining a processed film by the method of claim 3, 4 or 5, using a measuring scale in accordance with claim 6, the apparatus comprising detectors about the opposed portions of the said regions of contrasting colours and arranged along the lengths of the white or black regions, and comparision means for comparing the outputs of the detectors with a reference to provide an indication of the relative widths of the black or white regions.

8. A measuring scale substantially as herein described with reference to Figure 1 or Figure 2 of the drawings.